FinInteract (Frontier Pilot): leaderboard

Metric: Accuracy (%; the stratified 50-instance pilot subset run for the cross-vendor panel; ReAct agent in the standard full-interaction mode (A+S+I): it may search the filing corpus, ask the simulated user (GPT-5, answering yes, no or I don't know from the hidden disambiguating context) yes-or-no questions, and answer; a final answer counts as correct only if it matches the intended interpretation, graded by GPT-4o-mini with a 1% numeric tolerance). Source: arxiv.org. Saturation forecast: Around 2029. 7 models tracked.

Top models

#ModelScore
1Claude Sonnet 544
2GLM-5.240
3Kimi K334
4Grok 4.532
5GPT-5.6 Sol30
6Gemini 3.5 Flash30
7DeepSeek V4 Flash24

Interactive version: theaggregate.ai/benchmark?slug=fininteract-frontier-pilot · How It Works · Data refreshed daily, snapshot 2026-09-26.